Abstract
Mutation signatures in cancer genomes reflect endogenous and exogenous mutational processes, offering insights into tumour etiology, features for prognostic and biologic stratification and vulnerabilities to be exploited therapeutically. We present a novel machine learning formalism for improved signature inference, based on multi-modal correlated topic models (MMCTM) which can at once infer signatures from both single nucleotide and structural variation counts derived from cancer genome sequencing data. We exemplify the utility of our approach on two hormone driven, DNA repair deficient cancers: breast and ovary (n=755 cases total). Our results illuminate a new age-associated structural variation signature in breast cancer, and an independently identified substructure within homologous recombination deficient (HRD) tumours in breast and ovarian cancer. Together, our study emphasizes the importance of integrating multiple mutation modes for signature discovery and patient stratification, with biological and clinical implications for DNA repair deficient cancers.
